Xinming Chen, Pingwu Zhao, Yunsheng He, Kun Huang, Pan Zhao, Fengwan Liao, Yang Liu
{"title":"Development and validation of survival nomograms for patients with anaplastic thyroid carcinoma: a SEER program-based study.","authors":"Xinming Chen, Pingwu Zhao, Yunsheng He, Kun Huang, Pan Zhao, Fengwan Liao, Yang Liu","doi":"10.1007/s12672-024-01537-5","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>We aimed to study the prognostic risk factors affecting patients with anaplastic thyroid carcinoma (ATC), develop a clinical prognostic model, and assess patient survival outcomes.</p><p><strong>Methods: </strong>Patients with anaplastic thyroid carcinoma from 2000 to 2019 were selected from the Surveillance, Epidemiology, and End Results (SEER) Program to extract the clinical variables used for analysis. The dataset was divided into training (70%) and validation (30%) sets based on a 7:3 ratio. Univariate and LASSO regression analyses were performed on clinical variables from the training set to identify independent prognostic factors. Independent prognostic factors were determined by Univariate and lasso regression according to the clinical variables of the training set, and a nomogram model was established to construct a prognostic model based on the contribution degree of the predictors. The prognostic model was evaluated and internally verified by C-index, ROC curve and calibration curve.</p><p><strong>Results: </strong>A total of 713 ATC patients were included in the SEER database. LASSO regression results indicated that age, marital status, race, tumor size, whether the primary lesion was limited to the thyroid gland, surgery, radiotherapy and chemotherapy, were associated with overall survival(OS) prognosis of ATC, and were used to construct nomograms. In the training cohort, the OS nomogram's C-index was 0.708 (95% CI 0.672-0.745); in the internal validation cohort, the C-index was 0.677 (95% CI 0.620-0.735). ROC curves demonstrated that the OS nomogram exhibits excellent predictive accuracy and discriminative ability. Calibration curves indicated strong consistency between the OS nomogram's predicted survival rates and actual survival rates.</p><p><strong>Conclusions: </strong>We established a survival prediction model for ATC, which can assist clinicians in assessing patient prognosis and making personalized treatment decisions.</p>","PeriodicalId":11148,"journal":{"name":"Discover. Oncology","volume":null,"pages":null},"PeriodicalIF":2.8000,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11561200/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Discover. Oncology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s12672-024-01537-5","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Background: We aimed to study the prognostic risk factors affecting patients with anaplastic thyroid carcinoma (ATC), develop a clinical prognostic model, and assess patient survival outcomes.
Methods: Patients with anaplastic thyroid carcinoma from 2000 to 2019 were selected from the Surveillance, Epidemiology, and End Results (SEER) Program to extract the clinical variables used for analysis. The dataset was divided into training (70%) and validation (30%) sets based on a 7:3 ratio. Univariate and LASSO regression analyses were performed on clinical variables from the training set to identify independent prognostic factors. Independent prognostic factors were determined by Univariate and lasso regression according to the clinical variables of the training set, and a nomogram model was established to construct a prognostic model based on the contribution degree of the predictors. The prognostic model was evaluated and internally verified by C-index, ROC curve and calibration curve.
Results: A total of 713 ATC patients were included in the SEER database. LASSO regression results indicated that age, marital status, race, tumor size, whether the primary lesion was limited to the thyroid gland, surgery, radiotherapy and chemotherapy, were associated with overall survival(OS) prognosis of ATC, and were used to construct nomograms. In the training cohort, the OS nomogram's C-index was 0.708 (95% CI 0.672-0.745); in the internal validation cohort, the C-index was 0.677 (95% CI 0.620-0.735). ROC curves demonstrated that the OS nomogram exhibits excellent predictive accuracy and discriminative ability. Calibration curves indicated strong consistency between the OS nomogram's predicted survival rates and actual survival rates.
Conclusions: We established a survival prediction model for ATC, which can assist clinicians in assessing patient prognosis and making personalized treatment decisions.
背景我们旨在研究影响甲状腺无节细胞癌（ATC）患者预后的风险因素，建立临床预后模型，并评估患者的生存结果：从监测、流行病学和最终结果（SEER）项目中选取2000年至2019年的甲状腺无节细胞癌患者，提取用于分析的临床变量。数据集按 7:3 的比例分为训练集（70%）和验证集（30%）。对训练集的临床变量进行单变量和LASSO回归分析，以确定独立的预后因素。根据训练集的临床变量，通过单变量和拉索回归确定了独立的预后因素，并建立了一个提名图模型，根据预测因素的贡献程度构建预后模型。通过C指数、ROC曲线和校准曲线对预后模型进行评估和内部验证：SEER数据库共收录了713名ATC患者。LASSO回归结果表明，年龄、婚姻状况、种族、肿瘤大小、原发病灶是否局限于甲状腺、手术、放疗和化疗与ATC的总生存期（OS）预后相关，并被用于构建提名图。在训练队列中，OS提名图的C指数为0.708（95% CI 0.672-0.745）；在内部验证队列中，C指数为0.677（95% CI 0.620-0.735）。ROC曲线显示，OS提名图具有极佳的预测准确性和鉴别能力。校准曲线表明，OS提名图预测的生存率与实际生存率之间具有很强的一致性：我们建立了 ATC 的生存预测模型，它可以帮助临床医生评估患者的预后并做出个性化的治疗决策。
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
